4. RESULTS AND DISCUSSION OF PAPERS
4.1 P APER I
Bell RF, Sivertsen Å, Mowinckel P, Vindenes H. A bilateral clinical model for the study of acute and chronic pain after breast-reduction surgery. Acta Anaesthesiol Scand 2001; 45 (5):576-582
Results
VAS pain intensity scores: the model demonstrated a clear difference between lidocaine and placebo-treated breasts. The sum of VAS scores for pain intensity was significantly lower in the lidocaine group than in the placebo group for the entire registration period of 10 hrs after surgery. Regarding chronic postoperative pain there was no pain on testing at 6 months, however 3 patients reported ongoing periodic pain. There was no difference between lidocaine and placebo-treated breasts at 6 months.
Quantitative sensory testing: Five patients exhibited large changes in temperature thresholds (±5°C) 11 days after surgery, with no difference between lidocaine and saline-treated
breasts. Three of these patients reported ongoing periodic pain at six months, and exhibited significant thermal threshold changes, with no difference between lidocaine and placebo-treated breasts.
Discussion
Chronic pain after surgery is common120. Postoperative pain intensity has been shown to be a predictive of chronic postoperative pain128129130, and it has been suggested that
preemptive analgesia may reduce the risk of chronic postoperative pain.This has been been demonstrated in several clinical trials 13138. The hypothesis of preemptive analgesia was initially received with great enthusiasm, however no major clinical benefits have as yet been documented 132. Most studies in postoperative pain have focused on pain relief in the early postoperative period, and not on the development of chronic postoperative pain.
The advantageous aspects of the model we describe are that:
1. It is bilateral, with the patient serving as her own control.
2. The patients undergo sensory assessment with QST, using a thermotester which delivers specific stimuli, testing temperature perception thresholds. QST is thought to be
especially valuable in detecting impaired small fibre function which may be a factor in the chronification of acute pain. Quantitative sensory testing systems have been specifically developed to assess and quantify sensory function in patients with
neurological symptoms. Perioperative testing with long term follow-up may conceivably contribute to a better understanding of the mechanisms involved in persistent
postoperative pain. A report by the Therapeutics and Technology Assessment
Subcommittee of the American Academy of Neurology in 2003 concluded that QST is a potentially useful tool for measuring sensory impairment for clinical and research studies133. In our study it was interesting to note that the three patients who reported ongoing pain at 6 months also exhibited marked thermal threshold changes.
3. It involves a long-term follow-up regarding pain and sensory changes. This is in accordance with recommendations for clinical trials in postoperative pain114123.
The limitation of the model is that it is restricted to breast interventions and to the use of a local agent, such as a local anaesthetic.
Dahl and Moiniche123 have extensively discussed the concept of preemptive analgesia and suggest that the major problem is not nociception, but the prevention of central sensitisation.
As suggested in the paper, investigation of longer-lasting local anaesthetic or continuous application would be of interest. Prevention of central sensitisation in the clinical situation may be difficult, and Dahl et al122 123 recommend that future trials should investigate combinations of different classes of drug, such as ketamine, dextromethorphan or
gabapentin, all of which have demonstrated anti-hyperalgesic potential in clinical trials of postoperative pain. It would be interesting to test the effect of ketamine in this model since it is thought to have a local effect via peripheral NMDA receptors, and the expression of peripheral NMDA receptors has been shown to increase under conditions of inflammation17
18. An animal study has demonstrated that topical ketamine blocks topical morphine tolerance in mice134. An experimental study in healthy volunteers showed that local treatment with ketamine inhibits the development of secondary hyperalgesia in a burn model135. A randomised, controlled clinical study in postoperative pain also reported peripheral effects of ketamine136, however the group size in this trial was very small, less than 10 patients per group. A randomised placebo-controlled trial in patients with
neuropathic pain syndromes (diabetic neuropathy, postherpetic neuralgia, and postsurgical/
posttraumatic pain) found no difference between 1% topical ketamine and placebo137.
Since the publication of this paper new drugs, such as peripheral opioids, have been developed. The existence of an endogenous peripheral analgesic system is
well-documented22138. Using an in-vitro nerve-skin preparation from rats, peripherally delivered opioid (morphine) has recently been shown to directly inhibit the activity of cutaneous nociceptors under conditions of inflammation139. A peripheral opioid such as the -agonist frakefamide140 could therefore also be an interesting drug to investigate in this model.
Improving the model
The model could be applied to patients undergoing bilateral breast augmentation which is generally considered to be a more painful procedure than breast reduction and which has been shown be associated with chronic postsurgical pain130. This may further improve the sensitivity of the model. As suggested in the paper, time-consuming sensory testing with von Frey filaments, and VAS measurements on coughing could be eliminated, since these
provided no additional useful information. Assessment of pain relief in addition to pain intensity, and reporting of patient treatment preference would also improve the model.
Improving the paper- retrospective critical comments
The process of randomisation and allocation concealment should have been described. The number of patients should have been at least 10 (giving a minimum of 10 breasts in each group)88. No adverse effects were registered and this should have been stated, together with the method of assessment of adverse effects. The clinical significance of the demonstrated difference in pain scores versus the statistical significance, should also have been discussed.